RHELM (Long Context) - Mixed: leaderboard

Metric: Accuracy (%; 210 mixed dialogue-and-document questions, LLM judge, dialogue history plus external sources, full history in a 1M-token context). Source: arxiv.org. Saturation forecast: Around May 2028. 3 models tracked.

Top models

#ModelScore
1Gemini 2.5 Flash Lite13.1
2Qwen2.5-14B-Instruct-1M6.2
3GPT-4.1 Mini0.3

Interactive version: theaggregate.ai/benchmark?slug=rhelm-long-context-mixed · How It Works · Data refreshed daily, snapshot 2026-09-25.